[9fans] Fonts: PostScript 229 and 315 glyphes

[tlaronde@polynum.com]

Hello,

Since I'm moving to the 1.0 release of kerTeX, there is one area that I
need to tidy up: fonts.

The support for utf-8/runes will have to be put (later) in the
compilation, replacing the xord[] handling. This is the gestion of the
input.

In the meantime, the use of tcs(1) to convert to some encoding for TeX
input will do the job.

The next step in the process is the conversion of an input code to a TeX
command that can be a literal : a defined character. But TeX must have
the correct information for the correct glyph, and a dvi driver must put
afterwards the correct glyph in the correct place computed by TeX.

The correct way and the most flexible way to do the handling is to use
the "virtual fonts", the input code in TeX being mapped to some
sequences and/or some glyphes in whatever fonti(s) the virtual font is
combining.

For example, for "latin1" users, it can be far more easy to enter
directly the code for some non ASCII letter (even an "ASCII" accented
one) than to use the escaping commands.

So my goal is to tidy up this mess, and to provide virtual fonts mapping
to CMR or PostScript core (the core 35 fonts) and to "explicate"
(unfold/unhide) the process so that it can be relatively easy to adapt
to one's needs.

Adobe has provided not the definition of the fonts (the drawing) but the
metrics (what is needed by TeX to compute the layout).

There are two versions (see Redbook, E.5): standard (0229 glyphes) and
extended (0315 glyphes).

My intention is to provide the mapping by creating (for example) the
accented letters via a combination of glyphes from the 0229 glyphes (the
standard version).

Has anyone seen the "wild" use of the 0315 flavour? that has more than
just the accented glyphes.

--
        Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
                      http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C